Lin Hao, Lin Jinjin, Song Benteng, Chen Quansheng
School of Food and Biological Engineering, Jiangsu University, Zhenjiang 212013, China.
Foods. 2021 Mar 4;10(3):532. doi: 10.3390/foods10030532.
An olfactory visualization system conducts a qualitative or quantitative analysis of volatile organic compounds (VOCs) by utilizing the sensor array made of color sensitive dyes. The reaction chamber is important to the sensor array's sufficient and even exposure to VOCs. In the current work, a reaction chamber with an arc baffle embedded in the front of the air inlet for drainage effect was designed. The velocity of field and particle distribution of flow field in the reaction chamber was simulated by COMSOL Multiphysics. Through repeated simulation, the chamber achieved optimal result when the baffle curvature was 3.1 and the vertical distance between the baffle front end and the air inlet was 1.6 cm. Under the new reaction chamber, principal component analysis (PCA) and linear discriminant analysis (LDA) were employed to identify vinegar samples with different storage time through analyzing their VOCs. The LDA model achieved optimal performance when 8 principal components (PCs) were used, and the recognition rate was 95% in both training and prediction sets. The new reaction chamber could improve the stability and precision of an olfactory visualization system for VOCs analysis, and achieve the accurate differentiation and rapid discrimination of Zhenjiang vinegar with different storage time.
一种嗅觉可视化系统通过利用由颜色敏感染料制成的传感器阵列对挥发性有机化合物(VOCs)进行定性或定量分析。反应室对于传感器阵列充分且均匀地暴露于VOCs至关重要。在当前工作中,设计了一种反应室,其在进气口前方嵌入了弧形挡板以起到排水作用。利用COMSOL Multiphysics对反应室内的流场速度场和颗粒分布进行了模拟。通过反复模拟,当挡板曲率为3.1且挡板前端与进气口之间的垂直距离为1.6厘米时,该反应室取得了最佳效果。在新的反应室条件下,采用主成分分析(PCA)和线性判别分析(LDA)通过分析不同储存时间的镇江香醋样品的VOCs来识别它们。当使用8个主成分(PCs)时,LDA模型取得了最佳性能,训练集和预测集的识别率均为95%。新的反应室可以提高用于VOCs分析的嗅觉可视化系统的稳定性和精度,并实现对不同储存时间的镇江香醋的准确区分和快速鉴别。